By inperegelion
Deterministic, gated SDLC loop for full-stack JavaScript projects. Track-driven, Playwright-validated, security-reviewed, PR-completed.
Run the full cycle for one ticket — intake → implement → validate → review → pr → revalidate — dispatching each phase to a dedicated sub-agent. Gated; refuses to advance past a failing phase.
Implement the requirements in _/tracks/<TICKET>.md. Code, tests, mocks, seeds — and update the track's requirement statuses and journal as each lands.
Bootstrap agentic-sdlc for a project. Detects what it can (package manager, scripts, ticketing prefix), then walks the user through a short questionnaire for the decisions that can't be detected (validation strategy, spec convention, ticketing system). Writes _/sdlc-config.md.
Build _/tracks/<TICKET>.md — the single source of truth for one ticket. Verbatim ticket body, verbatim spec slice, acceptance criteria decomposed into testable requirements, YAML frontmatter that every later /agentic-sdlc:* command queries.
Walk every unresolved review comment on the PR, decide a verdict (Applied / Rejected / Deferred / Acknowledged / Stale / Duplicate), apply code changes when needed, and post a verdict-prefixed reply. Logs each thread to the track's journal.
Sub-agent dispatched by /agentic-sdlc:cycle for the implement phase. Plans against §6, implements requirements, runs the pipeline gate per requirement, returns a frontmatter delta with updated statuses and journal rows.
Sub-agent dispatched by /agentic-sdlc:cycle for the intake phase. Reads ticket + spec, decomposes ACs into testable requirements, returns a track frontmatter delta. Does not write files itself — the orchestrator persists.
Sub-agent dispatched by /agentic-sdlc:cycle for the PR creation phase. Re-runs the pipeline, commits via the commit-work skill, pushes the branch, opens the PR with a body built from the track frontmatter, and returns a frontmatter delta.
Sub-agent dispatched by /agentic-sdlc:cycle for the review phase. Coordinates three parallel reviewers (code, security, standards), consolidates findings into a CRITICAL/HIGH/MEDIUM/LOW report, and returns a frontmatter delta.
Sub-agent dispatched by /agentic-sdlc:cycle for the validate (and revalidate) phases. Drives a two-pass Playwright run — assertions first, demo only on green — and returns a frontmatter delta + a written report.
Use when the user asks Codex to run, continue, initialize, validate, review, or create a PR through the agentic-sdlc ticket-to-PR cycle.
Create high-quality git commits: review/stage intended changes, split into logical commits, write clear Conventional Commits messages, and run the project's pipeline gate before each commit. Use when committing within an agentic-sdlc cycle or whenever staging changes.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Ticket → merged PR — with a narrated demo video the stakeholders can actually watch.
A Claude Code plugin that runs a deterministic seven-phase loop: intake the ticket, implement against decomposed requirements, validate with Playwright and record a stakeholder .webm, run parallel code+security review, open the PR, address review comments, revalidate. Each phase has a verifiable gate — the cycle refuses to advance past a broken state.
ticket ──► intake ──► implement ──► validate ──► review ──► pr ──► revalidate ──► merge
▲ ▲ │ ▲ ▲ ▲
└───────────┴── gate ──────┤ └── gate ──┴── gate ──┘
└─► 🎬 narrated demo.webm
Every passing validate phase produces a .webm you can drop into a Slack thread, a Linear comment, or a stakeholder email. It's not a screenshot. It's not a test report. It's a narrated, slow-motion walkthrough of the feature behaving correctly:
validation.mode: standalone-playwright lets sdlc install and drive its own Playwright. Projects that never adopted E2E testing still get demos.force: true — if the UI is wrong, the run fails. The video is honest."I shipped a feature. Here's the 40-second video of it working." — what every PR description should be.
| # | Command | What it does | Gate |
|---|---|---|---|
| 0 | /agentic-sdlc:init | One-time setup. Detects package manager, scripts, ticketing prefix; wizard fills the rest. | Writes _/sdlc-config.md |
| 1 | /agentic-sdlc:intake | Pulls the ticket + spec, decomposes ACs into testable requirements. | Schema-valid track; every AC has ≥1 requirement |
| 2 | /agentic-sdlc:implement | Codes each requirement with tests + mocks, journals as it goes. | All requirements done; lint + typecheck + test + build green |
| 3 | /agentic-sdlc:validate | 🎬 Playwright two-pass: strict assertions report → narrated stakeholder demo .webm with overlays (only on green). | 0 assertion failures; no new console / 4xx-5xx errors |
| 4 | /agentic-sdlc:review | Three parallel sub-agents (code, security, standards) → CRITICAL/HIGH/MEDIUM/LOW report. | No unaddressed CRITICAL findings |
| 5 | /agentic-sdlc:pr | Pushes branch, opens PR with body built from the track frontmatter. | Pipeline green; PR URL written back |
| 5b | /agentic-sdlc:pr-comments | Walks every unresolved review thread, verdict-prefixes each reply (Applied / Rejected / Deferred / …). | Each thread logged to journal |
| 6 | /agentic-sdlc:revalidate | Re-runs validation against the post-review state; detects spec drift. | All requirements still pass; spec hash matches or drift ack'd |
| ∗ | /agentic-sdlc:cycle | Orchestrator — runs 1→6 in dedicated sub-agents, pausing between phases. | Stops at the first failed gate |
Every command reads and writes one file per ticket — _/tracks/<TICKET>.md — with YAML frontmatter as the contract and a freeform body for humans.
_/
├── sdlc-config.md project profile (frontmatter + agent notes)
├── tracks/PROJ-123.md track: requirements, status, journal
├── demo/PROJ-123.spec.mjs generated Playwright script
└── recordings/
├── PROJ-123.validation.md assertions report
├── PROJ-123.review.md consolidated review findings
├── PROJ-123.latest.webm stakeholder demo
└── …
npx claudepluginhub inperegelion/agentic-sdlc --plugin agentic-sdlcHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
AI-powered development tools for code review, research, design, and workflow automation.